1997-98 Knowledge and Concepts Examinations Grade 4 Results

Disaggregated Results: A Closer Look

Summary of Results

While Wisconsin fourth-grade students scored at higher levels in the subjects tested than fourth-grade students nationally, many Wisconsin students are not proficient or advanced. Analysis of these proficiency reports show:

Fewer fourth-grade students scored at or above the proficient level in enhanced language and mathematics than the other subject areas; overall performance was the highest in reading. The percent of students scoring at or above the proficient level in mathematics and science was substantially higher at grade 4 than at grades 8 or 10.
Girls scored higher than boys in reading and enhanced language; the performance of girls and boys in other subjects was about the same.
Proficiency levels, when reported by student mobility status, showed that students who have not been in the same school district for a full year had lower levels of proficiency than students who had been in the district for a full year. The educational benefits of staying in one district for a year or more appear less significant for disadvantaged student groups than for other students.
Large gaps exist in achievement between racial/ethnic minorities and white students and between economically disadvantaged and advantaged students. The differences in achievement across racial/ethnic groups are very large even among students who stayed in a single district for a full academic year and who are not economically disadvantaged.
Student groups demonstrating the lowest achievement levels also had the highest percentages of students who were not included in statewide testing. Many students with disabilities and limited-English proficiency who are required by law to be included in WSAS testing were not tested.
The percentage of nondisabled English-proficient students at grade 4 who took the tests was higher than at grade 8 or grade 10.

It is hoped that the information contained in this section and in this entire report will lead to a wider understanding of what we know, and what we don't know, about the achievement of student groups and schools in Wisconsin. It is also hoped that this analysis, and similar analyses conducted locally, will promote constructive, focused discussions involving staff, parents, and the community about ways to improve the learning of all students in each school.

Data Characteristics

All demographic and achievement data in this section are based on data collected by CTB/McGraw-Hill from school districts as part of the WSAS testing program. The data are as of October 1997 and are cross-sectional, not longitudinal.

Demographic data were provided by school districts for all students enrolled in grade 4 at the time of testing, October 1997. These data were not audited by CTB/McGraw-Hill or the Department of Public Instruction. Students demographic characteristics were coded for reporting purposes according to standardized definitions.

In order to reduce the data collection burden, default codes were used when the vast majority of students statewide were in a specific demographic category. For example, districts were only required to report a code for "students with disabilities." All students without a code were automatically coded "nondisabled." No default codes were used for race/ethnicity. Missing race/ethnicity codes were reported as "missing" or "N/A".

All achievement data are based on the results of standardized tests published by CTB/McGraw-Hill. These tests measure a wide range of important knowledge and skills but probably do not measure everything a local community believes is important for students to learn in school.

Students tested received a score in each of five subjects (reading, enhanced language, mathematics, science, and social studies). Each proficiency category ("minimal performance," "basic," "proficient," and "advanced") is associated with a range of scores. In all WSAS reports, percentages of students scoring in each of the four proficiency categories are based on percentages of students enrolled rather than percentages of students tested. Students who are not tested are included in a fifth category ("not tested on WSAS").

Due to space limitations, all graphs provided in this section of the report are for the WSAS grade 4 mathematics test unless otherwise indicated. Results for other subject areas suggest similar relationships between achievement and demographic characteristics of students.

Test Participation

MYTH: School achievement reports based on the results of students tested provide an accurate picture of overall achievement at the school.

FACT: This statement is true only if all or nearly all students are tested or students tested are comparable academically to students not tested. However, while many schools test all or nearly all students enrolled, many other schools do not. Student groups with the lowest achievement levels typically have the highest percentages of students who were not included in statewide testing.

School results based only on students tested would be higher for schools that do not test lower achieving students than they would be if the same schools tested all students. Students who are disadvantaged by mobility, economic status, disabilities or limited-English proficiency are tested at lower rates than other more advantaged students. Participation rates for racial/ethnic minorities are also lower than for white students.

Table 1 displays the percentages of students not taking WSAS tests at fourth grade by disadvantaged and nondisadvantaged student groups.

Table 1:
Percent of Students Not Taking WSAS Tests at Fourth Grade by Student Group

Table 1

*EEN stands for Exceptional Educational Needs. **Percentages vary by subject. *** Percentages vary by racial/ethnic group and by subject.

According to state law, districts may exclude two categories of students from WSAS testing. These are (1) EEN students with severe disabilities whose individualized education program (IEP) supports the need for exemption from WSAS tests and (2) LEP students who are at the lowest levels of English proficiency. Actual practices regarding exclusions vary widely across schools with many excluding all or nearly all students in one or both of these groups and many others excluding none. State law also permits parents to excuse their children from statewide tests, but the number of students affected statewide at Grade 4 is only 0.2%. State law requires that school districts administer the WSAS to all other students.

A three-week testing window is provided to allow local flexibility in scheduling and for administering makeup tests to students who are absent on test day. WSAS test participation rates vary across schools and school districts due in part to different local make-up testing practices.

As can be seen in the Figure 1, 8% of the EEN and 16% of the LEP students were not tested in Math and were not excluded or excused. Figure 1 also shows that 1.1% of nondisabled English-proficient students in grade 4 were not tested and not excused. The test participation rate for nondisabled English-proficient students at grade 4 was higher than the rate for grades 8 and 10.

Figure 1 graph

Achievement by Student Demographic Group

MYTH: Schools with higher overall student test results are more effective schools.

FACT: Schools vary in their effectiveness in addressing the needs of students in various demographic groups. Schools with higher overall test results often have fewer disadvantaged students and less racial/ethnic diversity than schools with lower overall results. Comparing results by demographic group provides a more complete picture of achievement at each school.

The educational task facing each school and each community is complicated by the specific needs and characteristics of students being served and by other factors. The relationships between student test scores and poverty, disability status, English proficiency, and racial/ethnic group are strong.

Figures 2 through 6 below illustrate the relationships for one subject, mathematics. Results for other subjects reveal similar relationships.

How to Read the Graphs

In the graphs labeled Figures 2 through Figure 6, the relationship between between test scores and economic status can be seen in the vertical distance between the dotted line for economically disadvantaged students in any given group and the solid line for students in the same group who are not economically disadvantaged. The greater the vertical distance, the stronger the relationship.

The relationships between test scores and disability status, limited-English proficiency, or race/ethnicity can be seen by the vertical distance between lines for the student groups being compared. If economically disadvantaged students are being compared, then the distance between the dotted lines is of interest. If students who are not economically disadvantaged are being compared, then the distance between the solid lines should be considered.

The strength of the association between test scores and mobility status of students who are not economically disadvantaged is seen in the slope of each solid line. For economically disadvantaged students the slope of the dotted lines should be considered. The steeper the line, the stronger the association.

Results for students who are limited-English-proficient (LEP) are displayed separately in Figures 3 and 6. Limited-English-proficient students are only 2.5% of students enrolled at grade 4 but are disproportionately represented in certain student groups. For example, 45% of all fourth-grade Asians are LEP. Inclusion of LEP students in the other graphs would have made interpretation of the relationships between groups more complicated.

Achievement by Disability and Economic Status

Note that, in Figure 2, results for EEN students are indicated by triangles, and results for nondisabled students are indicated by diamonds. Figure 2 shows that students who are not EEN (nondisabled) had higher rates of proficiency in math than EEN students regardless of economic or mobility status. The highest EEN group had 22% proficient or above in math. The lowest nondisabled group had 32% proficient or above in math.

Figure 2 graph

The association between rates of proficiency in math and economic status appears strongest for students enrolled in a district for a full year who are not EEN (vertical distance between lines is 30%). The relationship between rates of proficiency and economic status appears relatively weak for EEN students (vertical distance 11% for students in a district for a full year).

The association appears strong between mobility and achievement levels of students who are not EEN and not economically disadvantaged. Note that the solid line for this group has a steep slope. The association between mobility and achievement levels of economically disadvantaged students and EEN students appears relatively weak. Note that the lines for these groups are almost flat. The results also suggest that the benefits of a stable educational environment as measured by mobility may be far less for economically disadvantaged students, particularly those with disabilities, than for more advantaged students.

Achievement by English Proficiency and Economic Status

A strong association exists between rates of proficiency in math and economic status for English-proficient students, but the association is weak for LEP students. The association is also weak between rates of proficiency and mobility for economically disadvantaged limited-English-proficient students. See Figure 3.

Figure 3 graph

Achievement by Racial/Ethnic Group

Substantial differences in achievement exist across racial/ethnic groups and subject areas. African American students generally scored lower than students in other racial/ethnic groups. The scores for white students exceeded the scores for other racial/ethnic groups in all five subject areas.

Table 2 provides the percent of students in each racial/ethnic group that achieved scores at the proficient level or above in each of the five subject areas.

Table 2:
Proficiency Summaries By Racial/Ethnic Group

	Fourth Grade Proficient or Above
Subject Area	AmInd/AN	Asian/PI	African Amer	Hispanic	White	N/A
Reading	55%	47%	38%	47%	76%	57%
Language	23%	25%	13%	20%	46%	33%
Math	38%	36%	17%	28%	58%	44%
Science	50%	39%	21%	38%	72%	54%
Social Studies	48%	42%	23%	37%	69%	55%
Enrollment	919	1,998	6,518	2,149	50,312	1,286

While some of the differences in achievement across racial/ethnic groups are associated with economic status, mobility, or limited-English proficiency, Figure 4 indicates that these differences exist even among students who are not disadvantaged. The low proportions of certain racial/ethnic minorities scoring at the proficient level or above in mathematics do not appear to be explained by differences in mobility status, poverty, or English proficiency.

Figure 4 graph

The slopes of the lines in Figure 4 indicate possible effects of mobility on achievement of students in each one of the 10 student groups. The relationships between mobility and achievement are stronger for students who are not economically disadvantaged. Note that the slopes of the solid lines are generally steeper than the slopes of the dotted lines.

The data suggest that the benefits of a stable educational environment as measured by mobility may be less for economically disadvantaged students, particularly economically disadvantaged African Americans for whom the slope of the line is negative. Since 73% of fourth-grade African American students statewide are enrolled in Milwaukee Public Schools, analysis was done separately for this school district. The same pattern emerged.

The strength of the relationship between economic status and achievement can be seen for any given racial/ethnic group by looking at the vertical distance between the dotted line on Figure 4 for students in that group who are economically disadvantaged and the solid line for those in that group who are not economically disadvantaged. Among students who have been in a district for a full year, the relationship between economic status and achievement seems to be strongest for white students (vertical distance 22%) and weakest for African American students (vertical distance 11%).

Figure 5 is a simplified version of Figure 4. It includes only the results for the two racial/ethnic groups with the highest and lowest scores, whites and African Americans.

Figure 5 graph

Figure 6 summarizes results for LEP students by race/ethnicity and economic status.

Results are reported in Figure 6 when there are at least 25 students in a group. Results are not reported for many groups because only 2.5% of students in grade 4 statewide are LEP. 88% of LEP students are either Asian or Hispanic. 72% of LEP students are economically disadvantaged. Except in the case of economically disadvantaged LEP Asian and Hispanic students, the numbers of students in who were not in district for a full year were very small.

Figure 6 graph

LEP Asians outperformed Hispanics regardless of economic or mobility status. The relationship between achievement and mobility appears weak.

Implications of the Disaggregated Results

MYTH: Schools with higher proportions of disadvantaged students are expected to have lower overall student test results than schools with higher proportions of advantaged students.

FACT: The expectations for all students and all schools is the same. The long-term goal is for all students, including disadvantaged students, to progress to the proficient and advanced levels.

Some schools and communities have a long way to go towards the goal of having all students achieve the proficient level. This is especially true for many schools with large concentrations of disadvantaged students.

Educators, parents, and community members need to look at what is being done to improve learning for all student groups. What is being done to motivate unmotivated students and support their efforts to learn what's important? What programs are successful and should be continued or expanded and what programs are not and should be eliminated?

Possible ideas for change might come from schools that have higher than average achievement levels for disadvantaged groups.

For example, economically disadvantaged African American students at two Milwaukee schools, Clarke Street and Meir Elementary Schools, outperformed students statewide with 53% and 59% of students enrolled for a full academic year scoring at the proficient level or above in math. The statewide average was 52%.

Statewide, 29% of economically disadvantaged students scored at the proficient level or above in math. Economically disadvantaged African American students at 11 other Milwaukee schools outperformed economically disadvantaged students across the state.

Students with disabilities in seven school districts substantially outperformed students with disabilities statewide. On the math test, 40% percent or more of the students with disabilities enrolled in a single school for a full academic year scored at the proficient level or above in the Oconomowoc, Middleton-Cross Plains, Holmen, Eau Claire, Ashwaubenon, and Mequon-Thiensville school districts. Other districts and schools had higher percents at the proficient level but had fewer than 25 students with disabilities enrolled. The statewide average was 17% for this group.

For more information about these exam results go to:

For more complete information regarding statewide tests and other school performance indicators, see:

Wisconsin Student Assessment System (WSAS)

For questions about this information, contact dpistats@dpi.wi.gov (608) 267-3166

Last updated on 12/15/2011 4:00:36 PM